bank interest rate pass-through in the eurozone monetary policy transmission during the boom and...

8/2/2019 Bank Interest Rate Pass-through in the Eurozone Monetary Policy Transmission During the Boom and Since the Financial Crash

1/38

Bank interest rate pass-through in the eurozone :monetary policy transmission during the boom and

since the financial crash

First draft

This version: 2/10/2011

Christophe Blot1Fabien Labondance2

Abstract:

The aim of this paper is to understand how the financial crisis has affected the interestrate pass-through in the eurozone between market rates and bank interest rates. Weapplied a SUR-ECM model. This methodology allows testing for the homogeneity of thepass-through of the euro area countries. The main results of this investigation are thefollowing. First, not surprisingly, we show that the financial turmoil since august 2007 hasaffected drastically the interest rate passe-through in the eurozone. Second, the pass-through since the crisis is less complete that in the previous period studied. Third, theheterogeneity between the eurozone members has increased.


2/38

1. Introduction

Since the beginning of the stage three3 of the European monetary process, theEuropean Central Bank (ECB) has to implement a one size fits all monetarypolicy. It must be based of the monetary and economic conditions of the euroarea as a whole. Despite the economic and monetary integration process,member states banking system remains very specific. This heterogeneityinfluences drastically the monetary policy transmission. Regarding the OptimalCurrency Area (OCA) theory developed by Mundell (1961) and invoked byMongelli (2008) to deal with the question of the viability of the euro area. Onemain argument, called OCA meta-properties, is the homogeneity of monetarypolicy transmission process. Numerous studies have been dealing with thisissue (Angeloni, Kashyap, & Mojon, 2003) and the main conclusions is that itexists important cross-countries differences in the interest rate pass-through. But

it also appears that since 1999, this pass-through is quicker and tends to bemore homogenous between the euro area members.

In this paper, our purpose is not simply to study the influence of the monetaryunion on the pass-through, but rather to understand if the subprime crisis had

modified it. Our issue is to investigate if the interest rate pass-through in 11euro zone countries is influenced by the subprime crisis and how?

Banks play an important role in the transmission of monetary policy4, especiallyin the euro area where borrowers rely more heavily on the banking systems toraise funds. For a comparison, loans to the private sector granted by banks

amounted to 145 % of GDP in 2007 in the euro area against 63 % of GDP inthe United States. It is meaningful for central banks, since it involves that the

3 Stage one of Economic and Monetary Union (EMU) started in 1990 . It was characterized by


3/38

transmission of monetary policy strongly hinges on the speed and the size ofthe pass-through from policy-controlled interest rates to the bank interest rates. Itmay also be the source of heterogeneity in the transmission of the commonmonetary policy, since despite overall financial integration, credit markets aremuch less integrated in the euro area compared to stock, bond and moneymarkets (Jappelli and Pagano, 2008). Despite the convergence, which hasbeen realized in the run-up to EMU, heterogeneity is supposed to remainpervasive as differences in the banking systems across EMU countries stem not

only from differences in regulation or in the concentration of the bankingindustry but also from the local and private nature of the information. Theintrinsically asymmetric nature of information in credit markets is then a strongrestraint against complete convergence.

Since 1999, there were signs illustrating the potential role of an asymmetrictransmission in monetary policy. Differences in the developments in the housingmarkets were for example striking. Prices have more than doubled in nominalterms between 2000 and 2007 in France or in Spain whereas they onlyincreased by 3.5 % in Germany during the same period. At the same time, asurge in loans granted by the MFI (monetary and financial institutions) toresidents has been observed notably in Spain and Ireland whereas it was much

more muted in France and in Italy (chart 1). It even decreased in Germanywhere it went from 132 % of GDP in 2000 to 115 % in 2007. Thesedifferences inevitably raise a first question that will be dealt by this paper: arethese cyclical differences, at least partly, due to monetary policy? As banksplay a central role in the transmission of monetary policy, this issue will bedealt by focusing on the bank interest rate pass-through. Following, Sorensenand Werner (2006), we estimate a SUR-ECM model to test the heterogeneity of

the transmission of monetary policy across EMU countries.


4/38

Chart 1 : Loans to residents in % of GDP

50

100

150

200

250

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

Spain Germany

France Italy

Irelande Netherlands

Source: Eurostat

Since August 2007, the world has been through one of the most dramaticfinancial turmoil. The origins of the crisis may be found in the US subprimemarket but it then sprawled over the whole financial system. The panic gainedthe banking system since banks were directly or indirectly the final holders oftoxic assets. The financial net worth of borrowers and lenders have sharply

decreased, which could potentially impact on the transmission of monetarypolicy. The literature of the credit channels highlights the importance ofasymmetric information and agency problems which may have preciselybecome more acute with the deep financial shock. A second issue of this paperis then to investigate whether the financial shock may have triggered a break in


5/38

crisis is less complete that in the previous period studied. Third, theheterogeneity between the eurozone members has increased.

The remainder of the paper is organized as follows. Section 2 provides a shortreview on the literature related to the bank interest pass-though. This section willlead us to formulate research questions which will structure our results. Dataand econometric methodology are presented in section 3 whereas results are

detailed in the section 4. And finally we will conclude.

2. Overview of the literature and research questions

Numerous studies have been conducted on the subject of monetarytransmission pass-through in the euro area, but none, to our knowledge, sincethe subprime crisis. The vast literature on the bank interest rate pass-through hasrelied on two closed approaches. The first approach specifically deals with theissue of the monetary policy transmission channels5. ECB (2009) remindsindeed that policy-controlled interest rates are first transmitted through money

market and bond rates. The changes in these market rates are then passed-through the retail bank interest rates. It must be noticed that the degree of thispass-through is itself related to the monetary policy stance through its impact onthe values of collateral and on the credit risk assessment made by banks andthrough the effects of monetary policy on the health of the banking system. Thesecond approach is interested in the price-setting behavior of banks. Ithighlights the issues of monopolistic competition in the banking industry.

This paper is clearly oriented on monetary policy transmission since we aim atanalyzing the degree of heterogeneity among eurozone countries in thetransmission of the common monetary policy and a possible change in the


6/38

which will be less than one if banks have a market power. Papers interested inthe competition issues will use market rates money market or bond marketrates - at different maturities as the exogenous variables whereas theexogenous variable (mri,t) is a variable closely related with policy-controlledinterest rates when the focus is on the transmission of monetary policy. Thepolicy-controlled interest rates cannot indeed be used directly as they changeonly infrequently (de Bondt, 2005). The exogenous variable is therefore amoney market rate, which may be considered as a proxy for the policy rate

(Kwapil and Scharler, 2006). Concerning euro area, the EONIA (Euroovernight index average) rate may be the best indicator of monetary policystance as it is the rate that the ECB tries to influence through its refinancingoperations and through the marginal facilities (de Bondt, 2005). But EURIBOR6rates could also be alternative indicators since short-term contracts aresometimes indexed on these rates and since de Bondt (2005) has showed thatmonetary policy fully controls money market rates up to three months. In normal

times, EONIA and EURIBOR rates move fairly together but with the financialmarket turbulences, this relationship has been impaired. The empirical analysisprovided in this paper will consider the EONIA rate as the rate which is themost closely related to the ECB policy-rate. A robustness analysis will yet focuson the 3-month EURIBOR rate.

Starting from this simple pricing equation, empirical analysis have mainlyfollowed two methods for estimating the pass-through from market rates to bankinterest rates. Sander and Kleimeier (2004) for euro area Member states andde Bondt (2002 and 2005) at the euro area level estimated VAR models. Themain advantage of this approach relies on the simulation of impulse-reactionfunctions. The second and widely used approach consists in estimating error-

correction models7 (univariate or VECM). Equation (1) is supposed to be a long-term relationship and the estimated ECM model will show the adjustmentdynamics of the bank interest rates towards the stationary equilibrium. By thisway, we may appraise separately the short-term and the long-term pass-throughin the following estimated equation:


7/38

Finally, Sorensen and Werner (2006) provided a new and stimulatingapproach by estimating a SUR-ECM model. The main advantage of this panelapproach is that it allows testing for the homogeneity of the pass-through ineuro area countries.

A summary of the main results of this literature is provided in appendix I. Noclear conclusion emerges concerning the degree of the pass-through in theeurozone since analyses are based on different empirical models and areapplied to different countries as well as different sample periods. But, generalconclusions relative to lending rates may nevertheless be drawn. First, it isgenerally shown that lending rates are sticky in the short run; immediate pass-through are less than one. This sluggishness may come from customer switching

costs or menu costs leading to short-term nominal rigidities. In the long run,pass-through are higher and may be complete or not according to the countryand the credit market which is considered. It is argued that theseheterogeneities in the degree of pass-through are related to the legal andfinancial structures8. It is for example assumed that higher competition in thebanking system through indicators such as size of banks and concentration ofthe banking system - or from external finance related to the availability ofnonbank sources of finance - would increase the pass-through (Mojon, 2001). Affinito and Farabullini(2006) also highlight the importance of individualbanks characteristics such as the health of banks - liquidity and capitalizationsituation - as well as the way banks refinance their lending activity9 would alsoimpair the transmission of monetary policy. Thereby, differences amongeurozone Member states would be due to the heterogeneity of the national

banking systems.

It must also be stressed that the empirical results depend precisely on themodels which are estimated. The monetary policy and the cost-of-funds


8/38

complete pass-through from the short-term money market rate to the long-termgovernment rate, was not rejected by the data for the four largest Euro areacountries.

Finally, the pass-through may also change over time. First, it has often beenargued that the pass-through depends on the interest rate cycle. Concerningloan rates, it is expected to be faster when policy-controlled rates are

increasing than when they are decreasing. These cyclical asymmetries havebeen confirmed by Mojon (2001) and Sander and Kleimeier (2004).Otherwise, as the degree of pass-through hinges on the architecture of thefinancial system, it is subject to structural breaks, especially in the euro areawhere the transition period in the run up to EMU may have led to aconvergence of the legal and financial structures. The presence of one breakhas either been postulated it was then supposed to occur in 199910 - orendogenously determined. Sander and Kleimeier (2004) estimated a supremumF (supF) test for the monetary approach model and found that breaksgenerally occurred before the adoption of the single currency. For example,breaks occurred in 1994 and 1995 in the different Belgian credit markets. Itwas later for Germany since breaks were identified between 1996 and 2000.They also show that the size and the speed11 of the pass-through have

increased after the breaks implying a convergence process in the euro area.Based on beta and sigma convergence analyses, Vajanne (2007) confirmedan increased integration. Finally, a recent analysis carried by Marotta (2009)found that several breaks in the long-run pass-through have occurred during thetransition period. It is shown that the long-run pass-through has decreased in thelast period after the last break except for France and Ireland. Theadjustment speed has increased in all EMU countries but Portugal.

Firstly, in this paper, we will also consider the possibility that a break occurredin July 2007 because of the financial turmoil (Melvin & Taylor, 2009). Thereasons underlying such a break in the transmission channels are related to the


9/38

the bank loan market that monetary policy becomes less effective when theborrowers net worth is decreasing. The credit supply curve becomes steeperand the pass-though of cuts in the policy-controlled interest rate is weaker. Butwith the subprime crisis, the financial situation of the lenders the bankingsystem must be taken into account since it has incurred severe losses andfaced a stronger capital constraint. The lending channel precisely stresses thatthe health of banks influences the transmission of monetary policy. It can be firstargued that that the effects of monetary policy may be smaller when banks are

constrained by regulatory requirements. Even if monetary policy is eased,banks cannot expand credits since they can hardly raise new equity. But at thesame time, van den Heuvel (2002) argued that an expansionary monetarypolicy will alleviate the capital constraint by improving bank profits and willthen become more efficient. The consequences of the finance turmoil on thebank interest rate pass-through should then be tested. The first research questionwould then be the following:

1) Does the financial crisis affect the pass-through in the euro area?

With this question, the idea is to show that this crisis represent a structuralbreak for the interest rate pass-through in the euro area. The determination ofsuch structural breaks in the period before the crisis is often decidedexogenously with the beginning of the European Central Bank (ECB) in 1999

(De Bondt, Mojon, & Valla, 2005; Coffinet, 2005). But other studies also findstructural breaks endogenously. This the case of Sander & Kleimeier (2004)who determine a single break endogenously for each european countriesanalysed. They show that structural breaks in the long-run relationship betweenmarket rates and retail rates generally occur before 1999. Marotta (2009)extended these analysis in postuling the exitence of not one but multipleunknown breaks determined endogenously. He found some multiple structural

breaks for some countries indicating that national banking system are adjustingprogressively to the new monetary policy regime. Nevertheless, none studies, toour knowledge, have tested the impact of the subprime crisis which is certainlythe most important economic event that euro area has faced (Blot, Le Bayon,Lemoine, & Levasseur, 2009). The first issue in this research project is then to


10/38

may have worsened. Therefore, the aim of this paper is to tackle with the twofollowing issues:

2) Did the financial crisis modify the pass-through in the euro and did thefinancial crisis increase or mitigate the heterogeneity across EMU countries?

3. Data and Empirical Methodology

In this section, we provide details about the data used in this paper and aboutthe empirical methodology which is conducted. Our purpose is to estimate theinterest rate pass-through in the euro area countries and to be able to do somecross-country comparisons. We assume that these relationships are closelyrelated over time especially because they are affected with common monetary

shocks providing from the ECB decisions. But at the same time, we also want totest the heterogeneity in the transmission of monetary policy. We use a paneldata technique, the seemingly unrelated regression (SUR), to take into accountthe common structure of shocks as well as the possible heterogeneities in therelationship between the policy-controlled rate and bank interest rates. Thismethod has been recently extended by Kim (2004) or Moon and Peron (2005)

to integrate the dynamic relationship in SUR-ECM model.

3.1 Data

A harmonized database is used in this paper contrary to previous works exceptMarotta (2008) who precisely found that empirical results might be sensitive to

the choice of the database. Data are provided by the ECB and are called MFIinterest rate statistics. It covers those interest rates which are applied byresident monetary financial institutions (MFIs, i.e. "credit institutions") to euro-denominated deposits and loans to households and non-financial firms whichare residents of the euro area. More particularly, we focus on lending rates


11/38

3.2 Empirical methodology

Our analysis is conducted with interest rates which are potentially non-stationary time series. This must be taking into account because of the spuriousresults that can induce. Tests for unit roots using panel data are relativelyrecent13 and provide sometimes mitigated results. In this paper, we apply sixtests based on two different null hypotheses. The five first tests, i.e Levin Lin andChu, Breitung, Im Pesaran and Shin Fisher ADF and Fisher PP postulate theexistence of a unit root as the null hypothesis. The rejection of this hypothesisindicates stationarity. The last test, based on Hadri (2000) postulate as nullhypothesis the stationnarity. The rejection of the null hypothesis indicates a unitroot.

The results provided by these tests are presented in Table I. They clearly showthat the hypothesis of a unit-root is not rejected whereas, the null hypothesis ofstationarity in the case of the test performed by Hadri (2000) is rejected in allcases.

These results imply that we may work on spurious regressions. To deal with thispropriety of non stationarity, we postulate a cointegration relationship betweenbank and market rates. Since variables are cointegrated, we analyze their

relationship in an error correction model which allow us to quantify short termdynamics (with variables in first difference) and long term dynamics (withvariables in level). Several methodologies have been proposed. Stock &Watson (1993) proposed a dynamic OLS (DOLS) method which only estimatelong-term dynamic. Recently, Mark, Ogaki, & Sul (2005) extended the DOLSmethod to panel cointegration and thus defined a parametric method forestimating multiple cointegration regressions called the Dynamic SeeminglyUnrelated Regression (DSUR)14. DSUR methodology was applied by Moon &Perron (2005) for testing the purchasing power parity and it was also appliedto monetary policy transmission by Sorensen & Werner (2006). FollowingSorensen & Werner, we also estimate the equation (2) with this method. A SURis used because of the link between the estimated equations in their error terms


12/38

Table I : Unit root tests in panel data

Snf inf 1m Snf sup 1m immo conso

Ho: unit root, statistic and p-value

Levin, Lin,Chu

0,19

(0,57)

2.13

(0,98)

-0,33

(0,37)

-0,12

(0,45)

Breitung -0,99

(0,16)

1.10

(0.86)

-3,49

(0,00)

-1,27

(0,10)

Im, Pesaranand Shin

3,07

(0,99)

4.37

(1.00)

1,39

(0,91)

-0,35

(0,36)

Fisher-ADF 5,35

(0,99)

3.81

(1.00)

10,74

(0,97)

25,18

(0,19)

Fisher-PP 2,07

(1,00)

1.83

(1.00)

3,87

(1,00)

22,37

(0,32)

Ho: stationnarity, statistic and p-value

Hadri 8,46

(0,00)

9.55

(0,00)

8,15

(0,00)

7,27

(0,00)

More precisely, we estimate the interest rate pass-through between the market rate

( ) and the bank interest rate ( ) with the following equation


13/38

interest rates to avoid attracting riskier borrowers. reflects the speed ofadjustment to the long term equilibrium and measures the short-term

pass-through of market interest rates to retail bank interest rates.

This framework allows us to test the homogeneity of the coefficient across thecountries analyzed and thus compare and quantify the degree of heterogeneityin the interest rate pass-through in the short-term, in the long-term and we arealso able to compare the speed of adjustment.

3.3Break

Before, analyzing in depth the heterogeneity of the pass-through in the EMUcountries, we first consider the hypothesis that the transmission of the commonmonetary policy may have been impaired by the financial turmoil. This intuition

is illustrated by the Chart II where we clearly see that the monetary regime haschanged since august 2007.

Chart I1 : Evolution of the main refinancing operation rate, the eonia and the 3months Euribor since 2003

2

3

4

5

6


14/38

The hypothesis is tested with a Chow test. The date of the break is herepostulated since the beginning of the crisis is precisely defined. The test isapplied for each market, for the SUR-ECM model, for each individual countryand for the euro area as a whole. The Chow statistic indicates that the nullhypothesis of no break is clearly rejected for the SUR-ECM models (table A). Itis then necessary to disentangle the periods before and after the financial crisiswhen considering a multi-country model. When looking separately at each

individual country, the break is also confirmed in most of the cases andespecially for lending interest rates applied to non financial companies. Thenull hypothesis of no break cannot be rejected for Italy and only in the marketfor loans above 1 million Euros. Concerning households, the evidence is moremixed since the break is not significant in half of the countries. Neither forhousing market rates nor for consumer credit rates, the null of no break is

rejected in Germany, Italy, Belgium and Greece. Conversely, the break issignificant in the two households credit market in Spain, Ireland and Finland.But as, in the rest of the paper, the analysis mainly rests on the SUR-ECMmodels, we will postulate that a beak occurred and we will estimate the modelsfor the two sub-periods: the boom period from January 2003 to July 2007 andthe crisis period starting in August 2007.

Table II : CHOW Break test in 2007m08

Housingloans

Consumersloans

Businessloans for

NFC below 1million

Businessloans for

NFC above1 million

Germany 1.20 (0.32) 1.33 (0.26) 4.81 (0.00) 6.92 (0.00)France 3.31 (0.01) 1.39 (0.23) 2.24 (0.05) 2.38 (0.04)Italy 1.45 (0.21) 0.91 (0.49) 3.41 (0.00) 1.34 (0.25)

Spain 2.75 (0.02) 3.04 (0.01) 3.97 (0.00) 5.67 (0.00)Netherlands 2 10 (0 06) NA 3 31 (0 01) 2 95 (0 01)


15/38

Chow statistic (p-value). The null hypothesis is the absence of structural change. Thebreak is estimated in July 2007. For each individual country and for the euro area as a

whole, tests are implemented with univariate ECM estimated by OLS.

4. Results

The results presented here show the interest rate pass-through during the two periods:before and after the financial crisis which affected drastically both European

banking systems and the European monetary policy. We examine the results ofour estimation for interest rate pass-through affecting firms and households. Foreach interest rate pass-through well discuss the results for the short term pass-through, long term pass-through and the speed of adjustment (in appendix II,Table A1 to A4). We then examine some equality tests results to assess thedegree of heterogeneity within the eurozone (in appendix II, Table A5 to A48)

and finally, we propose some simulations which illustrate the impact of thefinancial crisis over the pass-through process (in appendix II, Charts A1 to A4).

4.1 Firms

In this analysis, we have selected two market rates which provide informationabout loan condition applied by monetary institutions to non financial

corporations, i.e. firms. We focus our work on the market rates for loans up toone million euro and over 1 million euro. These time series are available since2003 for all euro area countries, except for Greece for the loans over onemillion euro.

The table A1 shows the results for the interest rate pass-through between themarket rate and the rate apply to loans to non-financial corporations, up to 1million Euros. The results are divided in two sub-periods: before and since thefinancial crisis. First, it appears that for the pass-through in the short term, somewere ineffective in the boom period (France, Belgium, Portugal, Austria), but


16/38

states long-term pass-through before and after the beginning of the financial

crisis. We have combinations of equality or inequality of the pass-through between n member states of the eurozone. Before the crisis, weobtained 30 combinations where the pass-through were equal. Undoubtedly,the monetary policy process has become more heterogeneous and less effective(Charts A1) after, even though it seems that before the crisis the pass-throughexamined has reached a good level of homogeneity in the euro zone.

The table A2 shows the results for the interest rate pass-through between themarket rate and the rate applied to loans to non-financial corporations, over 1million Euros. The results are divided in two sub-periods: before and since thefinancial.Regarding the short-term and long term pass-through process we havebroadly the same results that the ones examined previously. The short-term pass-

through has increased since the financial crisis in all member states except one(Portugal) (Charts A2). But in the same time, the long-term pass-through hasdecreased and is complete for none of the member states examined. It remainsthat in most cases, the long-term pass-through is higher for loans over 1 million

Euros, which may generally be granted to bigger firms than the loans up to one

million Euros. It may simply come from an access to financial markets that is

facilitated for those firms. Higher competition between banks and markets would

increase the pass-through. On the subject of the speed of adjustment, it seems thatin general, the crisis have increased it except for two where it remains the same(Portugal, Austria). Nevertheless, as it is shown in the table A6, theheterogeneity has also increased in the long run after the crisis.

To sum up, the results highlight that the long-run pass-through has generally declined

since the financial crisis. Conversely, it seems that the short-term pass-though would

have increased. Banks would have anticipated that the ECB had engaged in a rapid

loosening of the monetary policy and expected then further policy rate cuts bringing

them to adjust downward the interest rates charged to firms. Otherwise, it also seems

that heterogeneity has increased with the financial turmoil.


17/38

increase (except in Italy). Charts A3 confirm us that the pass-through hasdecrease since the financial crisis. In this context, equality tests presented in the

table A7 are very interesting. Whereas during the boom period, 38 pass-through combinations were equal, it decreased to 12 since the beginning of thefinancial turmoil. The pass-through between market rates and mortgage loansrates was really affected by the crisis and worsened the heterogeneity betweenthe banking systems.

Estimations about the pass-through between money-market rates and consumerloans are less powerful. It seems that these interest rates are less impacted bythe monetary conditions than the others mentioned. It may be surprising sincethese loans are often short-term loans. Table A4 shows the results concerning theconsumer loans interest rates pass-through. In the short-term term, the pass-through seems to be inexistent and it appears very weak in the long term and

never complete (except in Finland). Nevertheless, the effects of the crisis are thesame than for the others pass-through: it weakened the long-term pass-through(Charts A4), increase the speed of adjustment and increase the heterogeneity inthe eurozone.

Table III recapitulates the main results for the four main interest rates examinedhere. We clearly see a before and an after subprime crisis in the average pass-through. It decreases after the crisis and the adjustment are more quicklyincorporate.

Table III : Average short-term pass-through (weighted averages), average longterm pass-through (weighted averages), average speed of adjustment (weightedaverages) and standard deviation in parentheses for the four rates examined,

during the boom and since the crisis


18/38

(0,14) (0,12) (0,19) (0,25) (0,17) (0,14)

0,23 0,48 0,80 0,70 -0,34 -0,40Loans tofirms upto 1Meuros

(0,16) (0,28) (0,11) (0,14) (0,13) (0,22)

0,45 0,76 0,96 0,81 -0,52 -0,88Loans tofirms

over1Meuros

(0,21) (0,18) (0,13) (0,09) (0,27) (0,21)

5. Conclusion

The aim of this paper is to understand how the financial crisis has affected theinterest rate pass-through in the eurozone between market rates and bankinterest rates. We applied a SUR-ECM model which is a panel method wherewe can estimate the pass-through between market rates and bank interest rates.This methodology allows testing for the homogeneity of the pass-through of theeuro area countries. The main results of this investigation are the following.First, not surprisingly, we show that the financial turmoil since august 2007 has

affected drastically the interest rate passe-through in the eurozone. Second, thepass-through since the crisis is less complete that in the previous period studied.Third, the heterogeneity between the eurozone members has increased.

From an economic policy view point, these results show us how the eurozone is

hard to manage in the way that is optimality regarding the OCA meta-property of homogeneity in the monetary transmission process is less effectivesince the financial crisis. A better economic governance seems appropriate todeal with these heterogeneities highlighted in this study.


19/38

Bibliography

Affinito, M., & Farabullini, F. (2006). An empirical analysis of nationaldifferences in the retail bank interest rates of the euro area. Banca d'ItaliaTermi di discussione series (589).

Angeloni, I., Kashyap, A., & Mojon, B. (2003). Monetary Policy Transmission

in the Euro Area. Cambridge.

Bernoth, K., & Von Hagen, J. (2004). Euribor futures market: efficiency and theimpact of ECB policy announcements. International Finance(7), 1-24.

Blot, C., & Cochard, M. (2008). L'nigme des exportations revisites. Que faut-il retenir des donnes de panel? Revue de l'OFCE(106), 67-100.

Blot, C., Creel, J., Labondance, F., Plane, M., & Timbeau, X. (2008). MonetaryPolicy in France and Germany: a case for different channels? workin paper-prepared for IMK/OFCE Conference in Berlin .

Blot, C., Le Bayon, S., Lemoine, M., & Levasseur, S. (2009). De la crisefinancire la crise conomique. une analyse comparative France-Etats-Unis.Revue de l'OFCE(110), 255-281.

Ceccheti, S. (1999, july). Legal structure, financial structure and the monetarypolicy transmission mechanism. FRBNY Economic Policy Review, 9-28.

Coffinet, J. (2005). Politique montaire unique et canal des taux d'intrt enFrance et dans la zone euro. Bulletin de la Banque de France, 136, 29-40.

Cottarelli, C., & Kourelis, A. (1994). Financial structure, bank lending ratesand the transmission mechanism of monetary policy. IMF Staff Papers, 41 (4),587-623.

De Bondt, G. (2005). Interest Rate Pass-Through: Empirical Results for the EuroArea. German Economic Review , 6 (1), 37-78.


20/38

ECB. (2009, August). Recent developments in the retail bank interest rate pass-through in the euro area. ECB Monthly Bulletin , 93-105.

ECB. (2008, july). The role of banks in the monetary policy transmissionmechanism. ECB Monthly Bulletin , 85-98.

Gambacorta, L. (2004). How Do Banks Set Interest Rates? NBER WorkingPaper Series (10295).

Graeve, D., De Jonghe, O., & Vander Vennet, R. (2004). The Determinants ofPass-Through of Market Conditions to Bank Retail Interest Rates in Belgium.National Bank of Belgiu Working Papers Series (47).

Heinemann, F., & Schler, M. (2002). Integration Benefits on EU retail CreditMarkets-Evidence from Interest Rate Pass-through. ZEW Discussion Paper(26).

Hofmann, B. (2006). EMU and the transmission of monetary policy: evidence

from business lending rates. Empirica, 33, 209-229.

Hurlin, C., & Mignon, V. (2007). Une synthse des tests de cointgration surdonnes de panel. Economie et prvision (178-179), 241-265.

Hurlin, C., & Mignon, V. (2005). Une synthse des tests de racine unitaire surdonnes de panel. Economie et prvision (169-170), 253-294.

Jappelli, T., & Pagano, M. (2008). Financial market integration under EMU.Economic Papers European Economy(312).

Kato, R., Ui, T., & Watanabe, T. (1999). Asymmetric effects of monetarypolicy: Japanese experience in the 1990s. Bank of Japan Working PaperSeries (2).

Kwapil, C., & Scharler, J. (2006). Limited pass-through from policy to retailinterest rates : empirical evidence and macroeconomic implications.sterreische National Bank Monetary Policy and the Economy, 26-36.

Mark, N., Ogaki, M., & Sul, D. (2005). Dynamic Seemingly Unrelatedf d


21/38

Mojon, B. (2000). Financial Structure and the Interest Rate Channel of ECBMonetary Policy. ECB Working Paper Series (40).

Mojon, B. (2001). Structures financires et canal des taux dintrt de lapolitique montaire dans la zone euro. Economie et Prvision (147), 89-113.

Mongelli, P. (2008). European Economic and Monetary Integration and theOptimum Currency Area Theory. Economic Papers, 302.

Moon, H. R., & Perron, B. (2005). Efficient Estimation of the SUR CointegrationReegression Model and Testing for Purchasing Power Parity. EconometricsReview(23), 293-323.

Mundell, R. (1961). A Theory of Optimum Currency Area.American EconomicReview, 51, pp. 657-665.

Sander, H., & Kleimeier, S. (2002). Asymmetric Adjustment of Commercial

Bank Interest Rates in the Euro Area: An Empirical Investigation into InterestRate Pass-Through. Kredit und Kapital(2), 161-192.

Sander, H., & Kleimeier, S. (2004). Convergence in the euro-zone retailbanking? What interest rate pass-through tells us about monetary policytransmission, competition and integration.Journal of International Money andFinance(23), 461-492.

Sorensen, C., & Werner, T. (2006). Bank Interest Rate Pass-Through in the EuroArea. ECB Working Paper Series (580).

Stock, J., & Watson, M. (1993). A Simple Estimator of Cointegration Vectors inHigher Order Integrated Systems. Econometrica, 61 (4), 783-820.

Thompson, S., Sul, D., & Bohl, M. (2002). Spatial Market Efficiency and PolicyRegime Change: Seemingly Unrelated Error Correction Model Estimation.American Journal of Agricultural Economics, 84 (4), 1042-1053.

Vajanne, L. (2007). Integration in the euro area retail banking markets-convergence of credit interest rates. Bank of Finland Research Discussion


22/38

22

Appendix 1: main findings in the literature

Studies data Timeperiod

Econometricapproach

Exogenous variable

Aggregation level

break extensions Main conclusions

Sorensen

& Werner(2006)

6 banking interest rate

(mortgage loans, consumerloans, short term and longterm loans to enterprises,current account deposits andtime deposits)

January

1999-June2004

DSUR Cost of

fundsapproach

European

individualcountry

Research of

the potentialexplanationsfor theobservedheterogeneity

High degree of

heterogeneity of the pass-through of market interestrates to bank interest ratesin the euro area.

They explain this result withthe degree of competitionin the states banking

sector.

De Bondt,Mojon, &

Valla(2005)

42 banking markets of theeuro area; 5 different retailbank segments (retail bankrates on short and long termloans to firms, mortgage

loans to households,consumer loans tohouseholds and timedeposits)

April1994-December2002

ecm Cost offundsapproach

10 euro areacountries andthe euro areaas a whole

1999 Retail rate depend on longterm market interest rate


23/38

23

De Bondt

(2005)

9 euro area retail bank

interest rates on depositsand loans

January

1996-may2001

Vecm eonia Euro area Hypothesis of

a change inthe interestrate pass-through sincetheintroductionof a common

monetarypolicy in1999.

The pass-through of official

interest rate to moneymarket interest rates up tothree months is completesince 1999. The immediatepass-through is incomplete,but the introduction of theeuro accelerates it.

Sander &Kleimeier,(2004)

10 different loan anddeposit rates

January1993-october2002

vecm Cost offundsapproachand moneymarket(monetarypolicyapproach)

10 euro areacountries

Determinationofbreaksforeachcountries,generallybefore

1999

Determinestructuralchanges ineuro-zonebankingduring therun-up tothe EMU

Structural breaks appearbefore the euro in 1999and the pass-through isquicker after these breaks.Distinction between the twoapproaches. Still marketimperfections which lead toheterogeneous pass-through.

Vajanne(2007)

6 different harmonized MFIinterest rate

January2003-december

Examine theconvergenceof retailbanking

Evidence for convergencein the interest rate.


24/38

24

2006 markets

Graeve, DeJonghe, &VanderVennet(2004)

14 products in the Belgianbanking market

January1993-december2002

ECM andpanel dataapproach

Cost of fund Micro level inBelgium

Determinantsofheterogeneity

Evidence of heterogeneitiesover loans and deposits. Itdepends positively on thematurity of the product.EMU didnt increase the

competition.vanLeuvensteijn, Sorensen,Bikker, &van Rixtel(2008)

6 banking interest rate(mortgage loans, consumerloans, short term and longterm loans to enterprises,current account deposits andtime deposits)

January1994-march2006

ECM andpanel dataapproach

Cost of fund 8 euro areacountries

Testing therelationshipbetweencompetitionand interestrate pass-

through

Bank interest rates in morecompetitive marketsrespond more strongly tochanges in market interestrates

Marotta(2009)

NRIR short term businesslending rates

January1993-september2003

ecm Cost of fund 12 euro areacountries

Severalbreaksaround1999

Implementinga search ofmultipleunknownbreaks

Multiple unknown structuralbreaks are found. Itsuggests caution inassociating structuralchanges to the introduction

of the euro. Bankingsystem are adjustingprogressively and theestimated PT is on averageless effective during the last


25/38

25

break-free period.

Gambacorta (2004)

73 banks march1993-march2001

Ecm andpanel

Moneymarket

Micro level inItaly

Analyzing awide rangeof micro andmacroeconomic data

variables thathave aneffect onbank interestrates

Heterogeneity in thebanking rates pass-throughexists, but only in the shortrun.

Sander &Kleimeier

(2002)

Lending rates from the IMF January1985-

december1998

ecm Moneymarket

15 EUcountries

Monetary policy in theeuro area is conducted

under the conditions of anasymmetric EMU.

Heinemann& Schler(2002)

NRIR three lending rates andtwo deposit rates

1995m01-1990

m10

ecm Cost of fund 11 EUcountries

Incomplete financialintegration which leads toasymmetries in the PT

processDe Bondt(2002)

Euro area retail bank interestrates on the deposits andloans

1996m01-2001m05

ecm Cost of fund Euro area Introduction ofthe euro

Sub samplewith theintroductionof the euro

The immediate PT isincomplete but quickersince the euro


26/38

26

Mojon

(2000)

25 credit rates and 17

deposit rates

Ecm and

panel

Money

market

6 euro area

countries

Focus on

financialstructurewhichcontribute tonationalasymmetriesin the interest

rate channel

Heterogeneity in the PT

Coffinet(2005)

7 interest rates 1986m01-2003m09

ecm Moneymarket

Euro areaand France

1999 The PT is quicker since theeuro in France and in theeuro area but this is not ashift of monetary regime.

Di Lorenzo

& Marotta(2006)

Nrir 1993

m01-2004m02

ecm Cost of fund Italy and

Portugal

Two

breaksin1999

1999 EMU didnt strengthened

PT


27/38

Appendix I1: Empirical results

Table A1: PT for 11 euro area countries during the boom and the crisis period(eonia) for loans to non-financial corporations, up to 1 million Euros , short

term, long term and adjustment coefficient (wald stat)

Short-term Long-term Speed of Adjustment

boom crisis boom crisis boom crisis

0,2 0,59 0,78 0,69 -0,28 -0,57DE(2,79) (51,38) (8,18) (437) (11,19) (21,89)

0,1 0,32 0,68 0,64 -0,29 -0,17FR

(0,31) (15,78) (9,56) (30,59) (11,64) (8,62)

0,28 0,44 0,84 0,79 -0,44 -0,43IT (8,28) (22,68) (19,64) (77,68) (39,35) (53,75)

0,34 0,43 0,93 0,64 -0,37 -0,32ES(11,61) (23,33) (2,88) (136,4) (35,36) (42,34)

0,37 0,41 0,84 0,46 -0,29 -0,41NL(7,14) (14,21) (3,62) (426) (10,13) (22,31)

0,009 0,54 0,99 0,83 -0,18 -0,62BE(0,004) (22,47) (0,009) (72,37) (1,55) (32,62)

0,51 1,04 1.01 0,78 -0,5 -0,3IE(12,86) (42,24) (0,1) (18,64) (26,64) (6)

0,16 0,31 0,72 0,68 -0,63 -0,32PT(1,24) (4,02) (62,19) (35,89) (31,18) (11,37)

0,28 0,38 0,97 0,82 -0,33 -0,63AT(2,28) (8,51) (0,12) (34,6) (14,59) (36,61)

0,3 0,55 0,79 0,81 -0,51 -0,64FI

(3,55) (16,54) (11,44) (63,41) (16,45) (30,61)

0 51 1 16 0 79 0 988 0 51 0 08


28/38

Table A2: PT for 10 euro area countries during the boom and the crisis period(eonia) for loans to non-financial corporations, over 1 million Euros , short term,

long term and adjustment coefficient (wald stat)

Short-term Long-term Speed of Adjustment


0,28 0,69 0,91 0,72 -0,63 -1,19DE

(3,18) (28,36) (4,44) (510) (25,52) (60,72)

0,71 1,05 1,06 0,88 -0,4 -0,82

FR (10,02) (30,93) (0,36) (25,22) (15,79) (42,28)

0,31 0,64 0,89 0,90 -0,56 -0,6IT

(1,84) (15,07) (3,31) (11,38) (18,39) (34,5)

0,34 0,65 1,26 0,87 -0,15 -0,87ES

(3,78) (24,6) (0,91) (64,26) (1,39) (56,26)

0,72 0,96 0.9 0,86 -0,57 -1,02NL

(6,41) (56,61) (0,02) (120,36) (12,4) (43,18)0,5 0,66 1,13 0,93 -0,38 -0,95

BE(7,45) (33,34) (2,41) (38,43) (9,59) (55,79)

0,57 0,66 1,01 0,92 -0,604 -1,001IE

(6,68) (20,28) (0,04) (25,85) (13,71) (38,72)

0,95 0,41 0,85 0,67 -0,44 -0,49PT

(21,94) (2,26) (4,63) (35,35) (8,75) (9,03)

0,63 0,79 0,95 0,85 -0,78 -0,7AT

(17,48) (18,86) (3,67) (40,7) (27,76) (21,49)

0,62 0,79 0,97 0,88 -0,73 -0,88FI

(9,04) (21,21) (0,47) (35,09) (24,32) (37,21)Short-term: null hypothesis: coefficient = 0,Long-term: null hypothesis : coefficient = 1,

Adjustment : null hypothesis : coefficient = 0,

In bold Short-term: short-term coefficients are null at the level of 10 %,(no pass-through in the short-term)In bold long-term: long-term coefficients are equal to the unity atthe level of 10 % (the pass-through is completed on the long term)In bold speed of adjustment : coefficients are null at the level of 10 %


29/38

Table A3: PT for 11 euro area countries during the boom and the crisis period(eonia) for mortgage loans short term, long term and adjustment coefficient

(wald stat)

short-terme Long-terme Ajustement


0.07 0.12 0.61 0.33 -0.08 -0.32DE

(0.53) (5.02) (2.49) (1129.66) (5.43) (17.61)0.29 -0.03 0.58 0.33 -0.03 -0.17

FR(14.50) (0.79) (1.27) (412.94) (2.17) (24.64)

0.26 0.24 1.00 0.83 -0.37 -0.20IT

(9.64) (12.05) (0.00) (12.75) (33.16) (21.11)

0.45 0.00 1.00 0.80 -0.23 -0.38ES

(37.28) (0.00) (0.00) (52.71) (13.06) (44.65)

0.10 0.05 0.67 0.15 -0.12 -0.19NL

(0.78) (1.33) (4.53) (1325.48) (12.03) (22.65)

0.13 0.10 0.55 0.41 -0.10 -0.17BE

(2.09) (3.76) (12.84) (302.76) (12.19) (17.98)

0.13 0.08 0.90 0.60 -0.56 -0.52IE

(0.92) (0.40) (8.32) (300.86) (17.26) (21.19)

0.31 0.33 0.74 0.83 -0.42 -0.41PT(7.77) (7.17) (38.84) (35.06) (20.42) (31.25)

0.34 0.04 0.73 0.67 -0.15 -0.39AT

(3.55) (0.08) (2.22) (50.87) (7.40) (18.62)

0.13 0.26 0.93 0.79 -0.29 -0.54FI

(1.45) (4.83) (0.92) (85.15) (17.90) (52.37)

-0.03 0.03 0.45 0.32 -0.06 -0.17

GR (0.07) (0.12) (2.05) (145.30) (2.41) (6.81)Short-term: null hypothesis: coefficient = 0,Long-term: null hypothesis : coefficient = 1,

Adjustment : null hypothesis : coefficient = 0,In bold Short-term: short-term coefficients are null at the level of 10 %,(no pass-through in the short-term)In bold long-term: long-term coefficients are equal to the unity at the level of 10 % (the pass-through is completed on the


30/38

Table A4: PT for 10 euro area countries during the boom and the crisis period(eonia) for consumer loans short term, long term and adjustment coefficient(wald stat)

short-terme Long-terme Ajustement


0.02 0.07 0.06 0.25 -0.30 -0.69DE

(0.00) (0.21) (18.65) (682.47) (6.25) (27.00)-0.20 -0.13 0.45 0.27 -0.21 -0.35

FR(0.65) (2.19) (8.95) (460.13) (10.07) (30.91)

-0.37 -0.32 -0.02 0.15 -0.27 -0.51IT

(1.46) (4.68) (32.68) (722.57) (18.16) (17.06)

1.17 -0.00 0.73 0.15 -0.71 -0.45ES

(6.34) (0.00) (6.71) (81.68) (19.81) (14.90)

-0.27 -0.05 0.21 0.51 -0.35 -0.19BE

(0.61) (0.07) (17.54) (6.28) (7.37) (3.78)

-0.47 1.28 0.41 0.25 -0.54 -0.25IE

(1.48) (17.90) (25.78) (11.69) (25.37) (3.36)

-0.66 -0.02 0.56 -0.48 -0.30 -0.28PT

(2.70) (0.01) (5.28) (37.78) (17.39) (10.09)

0.20 0.12 0.75 0.59 -0.19 -0.35AT(3.75) (1.43) (7.09) (144.84) (10.99) (23.44)

0.01 0.29 1.36 0.66 -0.09 -0.53FI

(0.00) (4.72) (0.48) (174.37) (4.78) (22.62)

-0.18 -0.30 0.02 0.06 -0.10 -0.33GR

(0.16) (2.35) (2.17) (231.66) (3.59) (7.22)Short-term: null hypothesis: coefficient = 0,

Long-term: null hypothesis : coefficient = 1,Adjustment : null hypothesis : coefficient = 0,In bold Short-term: short-term coefficients are null at the level of 10 %,(no pass-through in the short-term)In bold long-term: long-term coefficients are equal to the unity at the level of 10 % (the pass-through is completed on thelong term)In bold speed of adjustment : coefficients are null at the level of 10 %


31/38

31

Table A5: loans to non-financial corporations, up to 1 million Euros, Equality test on the long term PT between 11 euro zonecountries during the boom (the bottom of the table) and since the crisis (top of the table); Chi-deux stat is mentioned and

when the cell is gray-colored it means that the PT between those two countries are equal

DE FR IT ES NL BE IE PT AT GR FI

DE 0,58 21,54 6,27 70,28 60,38 3,75 0,004 49,9 0,25 34,11

FR 0,81 5,03 0,15 8,47 9,87 3,42 0,25 7,96 0,33 8,25

IT 0,48 2,84 142 170 7,02 0,01 3,51 3,41 0,11 1,8ES 3,48 6,94 12,65 23,64 73,67 12,69 1,99 62,93 0,38 42,23

NL 0,45 2,08 0,008 1,04 219 38,46 15,87 253 0,78 157

BE 1,51 2,47 0,76 0,1 0,63 1,28 6,98 0,93 0,07 0,76

IE 10,52 9,39 14,44 2,86 5,35 0,03 2,26 0,43 0,12 0,4

PT 0,4 0,27 6,21 17,47 1,92 2,14 31,3 5,94 0,26 4,69

AT 2,01 3,85 1,34 0,25 0,5 0,14 1,55 5,52 0,08 0,01

GR 0,002 0,86 0,77 4,46 0,46 1,43 13,45 0,89 2,27 0,08

FI 3,73 6,13 3,46 0,36 1,86 0,004 0,29 8,13 0,31 5,2


32/38

32

Table A6: loans to non-financial corporations, over 1 million Euros, Equality test on the long term PT between 11 euro zonecountries during the boom (the bottom of the table) and since the crisis (top of the table); Chi-deux stat is mentioned and

when the cell is gray-colored it means that the PT between those two countries are equalDE FR IT ES NL BE IE PT AT FI

DE 49,42 46,31 101,9 103,83 283 236 0,67 40,53 77,91

FR 2,73 0,75 0,13 0,42 5,38 3,98 13,26 1,9 0,01

IT 0,25 3,74 1,22 1,53 1,19 1,09 13,59 2,89 0,45

ES 0,35 0,71 2,15 0,31 27,06 9,35 11,59 1,27 0,17

NL 1,32 0,47 1,92 1,05 19,38 12,88 11,72 0,58 0,62

BE 8,61 0,43 7,93 0,25 2,71 0,02 19,18 12,11 6,28

IE 3,35 0,21 3,83 0,87 0,9 2,21 19,13 9,74 5,45

PT0,79 3,27 0,23 2,46 2,58 10,61 5,39 9,68 11,47

AT 0,89 1,6 1,19 1,4 0,49 6,31 1,32 2,01 1,87

FI 1,76 0,88 1,62 1,19 0,09 3,96 0,56 2,57 0,2


33/38

33

Table A7: Mortgage loans Equality test on the long term PT between 10 euro zone countries during the boom (the bottom ofthe table) and since the crisis (top of the table); Chi-deux stat is mentioned and when the cell is gray-colored it means that

the PT between those two countries are equal

DE FR IT ES NL BE IE PT AT FI GR

DE 0.03 146.29 239.29 71.74 9.27 84.70 301.14 44.74 383.81 0.00

FR 0.01 58.44 163.92 21.20 2.80 44.41 137.86 49.43 139.16 0.02

IT 2.72 1.27 0.43 220.38 79.02 24.45 0.01 5.55 1.12 50.66

ES 2.75 1.37 0.00 610.89 154.29 38.66 1.05 6.51 0.22 70.46

NL 0.05 0.06 5.66 5.83 62.54 215.71 453.84 102.48 601.54 12.23

BE 0.06 0.00 13.32 13.57 0.44 31.18 153.33 19.83 174.94 1.94

IE 1.39 0.74 7.22 5.80 2.44 8.10 43.19 1.50 52.36 20.92

PT 0.28 0.20 36.46 27.69 0.27 2.56 13.52 9.80 1.34 57.83

AT 0.16 0.15 2.08 2.17 0.06 0.61 0.79 0.01 5.77 20.90

FI 1.91 0.88 1.37 0.81 2.97 7.54 0.35 7.32 1.15 77.03

GR 0.15 0.06 2.13 2.15 0.32 0.07 1.36 0.60 0.39 1.47


34/38

34

Table A8: consumer loans Equality test on the long term PT between 10 euro zone countries during the boom (the bottom ofthe table) and since the crisis (top of the table); Chi-deux stat is mentioned and when the cell is gray-colored it means that

the PT between those two countries are equal

DE FR IT ES BE IE PT AT FI GR

DE 0.28 6.26 1.45 1.80 0.00 7.90 101.77 154.23 19.68

FR 2.25 8.57 2.33 1.47 0.01 5.47 68.04 95.18 17.95

IT 0.09 4.23 0.00 3.21 0.21 11.63 93.86 163.95 9.00

ES 6.52 1.71 19.71 2.82 0.21 9.34 25.44 26.92 4.29

BE 0.31 0.75 0.87 5.20 0.84 0.02 0.20 0.64 7.96

IE 4.02 0.03 6.05 4.09 0.87 1.10 2.68 3.53 2.11

PT4.52 0.18 7.05 0.51 1.82 0.63 1.84 4.72 36.25

AT 9.48 2.16 17.72 0.04 7.84 5.96 0.90 3.83 73.68

FI 5.31 3.02 6.49 1.46 4.80 3.08 1.98 1.45 102.4

GR 0.00 0.38 0.00 1.04 0.08 0.37 0.69 1.23 2.59


35/38

35

Charts A1 : simulation of the effect of an Eonias impulse of 100 basis points on the interest rate for loans to firms up to 1million Euros, during the boom and since the crisis in 10 euro zone countries

0

20

40

60

80

100

0 1:01 0 1:07 0 2:0 1 0 2:07 0 3:0 1 0 3:07

BOOM_AT CRISIS_AT

0

20

40

60

80

100

0 1:01 0 1:0 7 0 2:01 0 2:0 7 0 3:01 0 3:07

BOOM_BE CRISIS_BE

0

20

40

60

80

100

0 1:0 1 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

BOOM_ES CRISIS_ES

0

20

40

60

80

100

0 1:01 0 1:0 7 0 2:01 0 2:07 0 3:01 0 3:07

BOOM_DE CRISIS_DE

0

20

40

60

80

100

0 1:01 0 1:07 0 2:0 1 0 2:07 0 3:0 1 0 3:07

BOOM_FIN CRISIS_FIN

0

20

40

60

80

100

0 1:01 0 1:0 7 0 2:01 0 2:0 7 0 3:01 0 3:07

BOOM_FR CRISIS_FR

0

20

40

60

80

100

0 1:0 1 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

BOOM_IE CRISIS_IE

0

20

40

60

80

100

0 1:01 0 1:0 7 0 2:01 0 2:07 0 3:01 0 3:07

BOOM_IT CRISIS_IT

0

20

40

60

80

100

0 1:01 0 1:07 0 2:0 1 0 2:07 0 3:0 1 0 3:07

BOOM_NL CRISIS_NL

0

20

40

60

80

100

0 1:01 0 1:0 7 0 2:01 0 2:0 7 0 3:01 0 3:07

BOOM_PT CRISIS_PT


36/38

36

Charts A2 : simulation of the effect of an Eonias impulse of 100 basis points on the interest rate for loans to firms over 1million Euros, during the boom and since the crisis in 10 euro zone countries

0

20

40

60

80

100

120

0 1:01 0 1:07 0 2:0 1 0 2:07 0 3:01 0 3:07

B O OM _ AT C R I SI S_ A T

0

20

40

60

80

100

120

0 1:0 1 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B O OM _ BE C R I SI S_ B E

0

20

40

60

80

100

120

0 1:01 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B OO M_ D E C R IS IS _D E

0

20

40

60

80

10 0

12 0

0 1:01 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B O OM _ ES C R I SIS _ ES

0

20

40

60

80

100

120

0 1:01 0 1:07 0 2:0 1 0 2:07 0 3:01 0 3:07

B O OM _ FIN C R I SIS _ FIN

0

20

40

60

80

100

120

0 1:0 1 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B OO M _F R C R I SIS _ FR

0

20

40

60

80

100

120

0 1:01 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B OO M _IE C R I SIS _ IE

0

20

40

60

80

10 0

12 0

0 1:01 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B O OM _ IT C R I SI S_ IT

0

20

40

60

80

100

120

0 1:01 0 1:07 0 2:0 1 0 2:07 0 3:01 0 3:07

B O OM _ N L C R IS IS _ NL

0

20

40

60

80

100

120

0 1:0 1 0 1:07 0 2:01 0 2:07 0 3:01 0 3:07

B O OM _ PT C R I SI S_ P T


37/38

37

Charts A3 : simulation of the effect of an Eonias impulse of 100 basis points on the interest rate for mortgage loans, duringthe boom and since the crisis in 11 euro zone countries

0

20

40

60

80

100

120

0 1:01 0 1:0 7 0 2:0 1 0 2:0 7 0 3:0 1 0 3:07

B OO M_ AT C R IS IS _A T

0

20

40

60

80

100

120

0 1:0 1 0 1:07 0 2:0 1 0 2:0 7 0 3:01 0 3:0 7

B OO M_ BE C R IS IS _B E

0

20

40

60

80

100

120

0 1:0 1 0 1:0 7 0 2:01 0 2:0 7 0 3:01 0 3:0 7

B OO M_ D E C R IS IS _D E

0

20

40

60

80

100

120

0 1:01 0 1:0 7 0 2:01 0 2:0 7 0 3:0 1 0 3:07

B OO M_ ES C R IS IS _E S

0

20

40

60

80

100

120

0 1:01 0 1:0 7 0 2:0 1 0 2:0 7 0 3:0 1 0 3:07

BOOM _ FIN CRISIS_ F IN

0

20

40

60

80

100

120

0 1:0 1 0 1:07 0 2:0 1 0 2:0 7 0 3:01 0 3:0 7

B OO M_ F R C R IS IS _F R

0

20

40

60

80

100

120

0 1:0 1 0 1:0 7 0 2:01 0 2:0 7 0 3:01 0 3:0 7

B OO M_ G R C R IS IS _G R

0

20

40

60

80

100

120

0 1:01 0 1:0 7 0 2:01 0 2:0 7 0 3:0 1 0 3:07

B OO M_ IE C R IS IS _IE

0

20

40

60

80

100

120

0 1:01 0 1:0 7 0 2:0 1 0 2:0 7 0 3:0 1 0 3:07

B OO M_ IT C R IS IS _I T

0

20

40

60

80

100

120

0 1:0 1 0 1:07 0 2:0 1 0 2:0 7 0 3:01 0 3:0 7

B OO M_ N L C R IS IS _N L

0

20

40

60

80

100

120

0 1:0 1 0 1:0 7 0 2:01 0 2:0 7 0 3:01 0 3:0 7

B OO M_ PT C R IS IS _P T


38/38

38

Charts A4 : simulation of the effect of an Eonias impulse of 100 basis points on the interest rate for consumer loans, duringthe boom and since the crisis in 10 euro zone countries

-2 0

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B OO M _A T C R I SIS _ AT

-20

0

20

40

60

80

100

120

0 1:01 01:07 02:01 02:07 03:01 03:07

B O OM _ BE C R I SI S_ B E

-20

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B OO M _D E C R IS IS _D E

-20

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B OO M_ E S C R I SIS _ ES

-2 0

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B O OM _ FIN C R I SIS _ FI N

-20

0

20

40

60

80

100

120

0 1:01 01:07 02:01 02:07 03:01 03:07

B OO M_ F R C R I SIS _ FR

-20

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B O OM _ GR C R IS IS _ GR

-20

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B OO M _IE C R I SI S_ IE

-2 0

0

20

40

60

80

100

120

01:01 01:07 02:01 02:07 03:01 03:07

B OO M _IT C R IS IS _IT

-20

0

20

40

60

80

100

120

0 1:01 01:07 02:01 02:07 03:01 03:07

B O OM _ PT C R I SI S_ P T

bank interest rate pass-through in the eurozone monetary policy transmission during the boom and...

Documents